After leaving the Fruit Machine emulation for a bit while 0.144 was put out the door I think it’s about due time I got back to looking at that. I’m effectively down a man on that front tho, because James who was providing some assistance is currently out of action due to real life goings on.
Also, as outlined below NeoGeo Multislot is in my plans, this shouldn’t actually be too difficult to implement now, just requires a bit of planning.
There are a couple of MESS-side issues I’ve noticed while going through systems with UltimateMAME which could do with some attention (not specific to my build, just general problems in MESS) For now I’m just waiting to see what the actual MAME/MESSdevs come up with as a solution as far as building a unified binary goes tho. If they are serious about doing a merge then any sensible solution would at least have the option to build an ‘UltimateMAME’ like target, so I’m going to bide my time on that for a while. I’ve shown that it can be done so the ball is very much in their court now as far as delivering anything as complete and functional as what I provided, but in an official capacity is concerned.
I _won’t_ be working on Raiden 2, Gaelco, HNG64, Later release PGM or anything along those lines, sorry.
I might do some more mini-reviews / game(s) of the week type things, pointing out some obscure, or forgotten games which are noteworthy for some reason or another. There are many interesting titles out there beyond MAME, and due to various factors (difficulty in using the systems without prior knowledge, vast size of software libraries, assumptions that old computers/consoles just weren’t as capable, lack of a popular MAME-like project for non-arcades, the number of closed source ‘best’ emulators which simply fail on modern platforms etc.) a fair number are being completely overlooked, or at most relegated to ‘best 100 games for xxx’ YouTube videos with little explanation as to why.
I’ll probably use my UltimateMAME builds, or MESS for the purpose of this, as while compatibility, and usability may be inferior to some of the standalone emulators people can actually depend on it to work as described regardless of their platform. It’s just a shame compatibility on some of the systems is so bad (PSX, Saturn, Jaguar etc.) of course being open software there is always the opportunity to improve it, although that requires people to actually know it exists, and one of the most evident things about UltimateMAME is that a lot of people apparently didn’t even realise what MESS was, so I can only guess it’s just as hidden from developers.
With the above in mind, one final thing I will say is that it MESS exists, and if you’re considering doing your own 100% accurate emulator for a platform please consider coding in MESS instead, and improving the MAME framework while you’re at it because one thing about the project(s) is they’re always going to be there and somebody is always going to be maintaining them, so if you care about things being preserved for the generations it’s a good place to do your work. It might seem like a silly ‘nothing’ project split off from MAME, and relegated to obscurity, but it really is just as important, if not more and we’re quickly approaching a period where most of the developers who are likely to be interested in these things are getting older, and moving on to other real life ventures which IMHO puts some systems at risk of never being emulated and preserved properly in an open, active project.
testing.. odd, it said there was one comment but now it still says there’s one comment
Ok,Haze,do you have gave up emulate arcade hardeware for Mame instead of Mess?
Hi Haze,
Two things.
First, do you have any feedback on if your builds “ignited” any positive reaction on the core MAME devs?
Second, indeed MESS is tough to fight some stand-alone emulation projects, but don’t say they fail on modern platform. Some yes. Most common historical systems though have VERY active emulation teams that make emulators that run perfectly on modern platforms.
WinUAE (and UAE) releases the sources. BY FAR the most complete, Amiga emulator.
VICE too. It is also VERY MUCH multiplatform.
Stella (for Atari 2600) is open source.
Same goes for good Sinclair and Amstrad emulators.
Yes, things are improving, projects like WinUAE will always have things to do, because there’s always something new going on with the Amiga to keep the proejcts active. I don’t think MAME/MESS would ever replace WinUAE due to the sheer flexibility it offers, although it would be good to see solid emulation of all the base components, and we’re a long way from that!
I question the Sinclair / Amstrad stuff tho, I struggled to find a decent replacement for things like RealSpectrum last time I tried, and the authors of that have even demanded it no longer be distributed for some strange reason.
For the Sega consoles the situation isn’t ideal, Kega still has a stranglehold on users, and is entirely closed. Even for the Playstation pSX isn’t _that_ good, the best Saturn emulators likewise. Even for some other systems you end up with situations where the emulation is considered ‘done’ and the development becomes inactive, which means if critical changes are needed to keep it running they’re not as likely to happen. A lot of people recommended Nestopia for NES for a long time, but the latest releases are causing sound issues on modern systems for people and the author is nowhere in sight. (I know there are other alternatives, but it’s still not an ideal situation for people who came to like that) Then you get the obscure Japanese systems, where emus pop up and fade away quickly, MESS is sometimes the only emulator for some of the platforms, and from what I can see most of the other emulators for other popular platforms (such as the X68K) are depending on things like Starscream for their CPU cores, which is automatically going to be bad for compatibility, working on them in MESS simply makes more sense.
Obviously drivers in MAME/MESS fall into disrepair too, usually if they were badly coded in the first place tho, although sometimes due to untested core changes, however they remain in an active project which is important IMHO.
In some cases, like the UAE one, and several other computers having alternatives is good because the systems are often complex and expandable. For other cases, mainly consoles, it’s just creating fragmentation, divided efforts, wasted efforts, multiple standards, and an overall messy situation with little focus on actually doing things correctly. It’s only in recent years with the Mess Software Lists we’ve even seen any effort to properly document console cartridges and the actual roms contained in them. Too much fragmentation makes it hard to know where to turn, having a solid base reference point in a single emulator, even if it then leads to exploring other emulators is better than not knowing where to start. Other emulators can always offer things MAME/MESS won’t because they simply don’t make sense if you’re trying to do things properly (higher resolution 3d rendering on things like the Playstation for example) but having something which does the basics right, for reference, is important.
As for positive feedback, nothing, as expected. Seems practically everybody wants to see this aside a few influential people on Mamedev, who are more likely to actively try and stop such a combined build from working than support it’s creation. From what I’ve been told their specific plans are very much tailored around avoiding it, even tho it would actually be easier to allow it.
As individuals, and as teams of people I feel we all have a responsibility to get things right for future generations, to provide high quality, accessible emulation, to provide a ‘go to’ point, a reference. To effectively do this the main project has to be open the new ideas, and open to anything which could improve the emulation, and that means it can’t be artificially limited to a small subset of the useful targets, such as arcade games. Running a lot of stuff in MESS really shows how weak the MAME emulation still is in areas, and that’s not acceptable.
Hey Haze, if so many people are for a combined MAME/MESS build why not fork and let the people decide which to continue with? (This isn’t a flame just asking).
I know there is problems having a 3rd MAME/MESS build but wouldn’t the ones that preferred this way specifically move to it? Might be just the kick in the pants the others need to change over?
Anyways, thanks for the interesting posts :)
As I said.. I don’t think a fork is worthwhile, it isn’t my intention to further fragment things, which is all a fork would achieve, and without the official support / ties to the development tree it would just be a burden to maintain.
People tend to stick with what’s official, because it’s official, not because it’s better. Development happens there too. Without it being official things would get broken in the compile all the time due to active devs simply not caring / not testing, and I’d have to pick up the pieces.
The majority of other non-official builds / forks have died out, or ended up on life support for similar reasons.
The actual team need to provide some way to do this as a real target, it wouldn’t be much work for them if they merge everything else and bring the build systems closer together, any reasons not to have such a target become purely political at that point.
If there are any bits of low hanging fruit you want me to work on, I don’t mind doing bits and pieces again now, but none of the big stuff, as I don’t have the time.
I tried posting on one of the FME sites to try to get some database help for you, but no takers on that front.
I _won’t_ be working on Raiden 2, Gaelco, HNG64, Later release PGM or anything along those lines, sorry.
NOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO
1- epsx is sh*t, psx is sh*t, Xebra is MAME-like (closed source). *Put your saturn emulator here* is shit. SSF is MAME-like (closed source). *Put your super nes emulator here* is shit. BSnes is the shit! (nigga from there has a nice coding style)
2- I never used Kega. Meka (hated the lack of vertical left bar emulation) + Gens always did it (for me), now MESS does.
3- *Waiting for MESS to emulate the 9999-in-1 “mini-games” (accurately)!!!*
4- It is not that MESS is “hidden”, it is just that an emulator to get “famous” must emulate king of fighters 2002, super mario world, zelda something and marvel vs capcom 2. [running at fullspeed (with sound) in 400 fps unthrottled]
5- Don’t worry, there will always be *lovable* ppl seeking to emulate old and forgotten things:)
6- F*** Raiden 2 (sry fans)
6- I realized that realised is a valid word in UK. (Thanks to Google)
7- Penguin Land is fyrweyriuwehr hard, gave up already!
8- HLSL is Windows only (if i am not wrong), MAME uses it. CUDA (nvidia) and the *i don’t know the name* (amd ati) could be used for drawing operations? If so, why MAME doesn’t use it?
Hello Haze… Yes there was a first post I made it but it`s not here, you deleted it? I will repeat….
I`ve worked on my own emulators (… I`ll spare you the details) but Mess has never been something I wanted to work on for only some of the reasons here
It`s the unloved offspring of Mame, the relationship is -~*strange*~-. There are terrible UI design issues, no user base and too many bullish egos surrounding it.
Being mindful any code I worked on would also need to work with arcade games which weren`t even compiled or linked to the project wasn`t a natural working environment. I could never understand why it was two projects……..
I knew about Mess but I avoided it
I understand the sentiment of your appeal but I`m sure you`ll also understand mine
^_^
Heads up from the FME world here.
Bud, least we are getting updates and results, far better than we have had in times past in the FME world.
Appreciate the work you are putting in.
Next time I am up in your world, beers are on me, which considering how close I am, is a possibility :D
I hate it that politics guide things in such a field! It is stupid. I mean can’t the rest let those people GO? If they want to stop developing, let them stop!
MAME doesn’t belong to anybody!
CGR: You forgot the Nintendo 64 emulators, they have to be up there with ePSXe. :)
Heitachi_73:
Oh, how could I? Apologizes my friend.
N64 emulators are shit!
and the only other capable CDI emulator is commercial, closed source..
The only 3DO emulator has some weird license which doesn’t really have agreeable terms at all (can’t really be used with anything else, renders anybody who even looks at the source to be in a position where they can’t even contribute to anything else)
The existing fruit machine emulators are kludgy hacks making no real attempts to document the software… the most popular of which already struggle to run properly on Win 7.
Don’t get me wrong, there probably isn’t a single system in MESS where the MESS emulation is better than the standalones, but if we’re working for the future then working in projects like MAME/MESS makes a lot more sense and any code gets vigorously tested against multiple systems giving you the best selection of test cases and the potential to write the best possible emulation.
But for that to happen people need to get involved, and to notice MESS, contribute to MESS, and improve MESS, not just start another standalone.
If you are using a Unix or Mac OS, MESS becomes a lot more attractive because for a lot of systems there are no alternatives…
Also the latest version of MESS is basically the only emulator to do LLE of Macintosh, including bus expansion and ethernet cards (you can use emulated Netscape and your internet connection to browse the web!), and to emulate the Avigo PDA. Emulation is also on par with standalone emus for many exotic systems (eastern europe or scandinavian 8/16bit computer and a few Jpn computers like the Sony SMC777)
Of course if you are only interested into console and games (or mainstream home computers) the situation is different and there is still a lot of space for improvements… but still improvements are happening all the time (e.g. a new guy, Huygens, has done marvels with the intv and is now fixing some lynx issues)!
Well yeah, for non windows your options are even more limited..
As I said, for a bunch of the obscure ones Mess has an advantage by default, it’s not being build of ancient cores like Starscream, and the Japanese mentality of ‘must HLE the entire bios / system so that free software can be used’ isn’t helping their cause as far as emulation goes at all.
It still seems pretty evident from the comments here, and on previous articles that having it all part of one project would be best tho, increased exposure, increased development opportunity, and more. I don’t get what the teams really have to lose by embracing something like that, at worst you see no change to the current levels of development. Regardless, a lot more people *see* what the MESS side of things can do.